arXiv subjects

Andrew Farley

Publications and source records attributed to Andrew Farley.

3 recordsLinked to original sources

arXiv2023

Diffusion Dataset Generation: Towards Closing the Sim2Real Gap for Pedestrian Detection

We propose a method that augments a simulated dataset using diffusion models to improve the performance of pedestrian detection in real-world data. The high cost of collecting and annotating data in the real-world has motivated the use of simulation platforms to create training datasets. While simulated data is inexpensive to collect and annotate, it unfortunately does not always closely match the distribution of real-world data, which is known as the sim2real gap. In this paper we propose a novel method of synthetic data creation meant to close the sim2real gap for the challenging pedestrian detection task. Our method uses a diffusion-based architecture to learn a real-world distribution which, once trained, is used to generate datasets. We mix this generated data with simulated data as a form of augmentation and show that training on a combination of generated and simulated data increases average precision by as much as 27.3% for pedestrian detection models in real-world data, compared against training on purely simulated data.

cs.CV↗

arXiv2022

Multiscale Crowd Counting and Localization By Multitask Point Supervision

We propose a multitask approach for crowd counting and person localization in a unified framework. As the detection and localization tasks are well-correlated and can be jointly tackled, our model benefits from a multitask solution by learning multiscale representations of encoded crowd images, and subsequently fusing them. In contrast to the relatively more popular density-based methods, our model uses point supervision to allow for crowd locations to be accurately identified. We test our model on two popular crowd counting datasets, ShanghaiTech A and B, and demonstrate that our method achieves strong results on both counting and localization tasks, with MSE measures of 110.7 and 15.0 for crowd counting and AP measures of 0.71 and 0.75 for localization, on ShanghaiTech A and B respectively. Our detailed ablation experiments show the impact of our multiscale approach as well as the effectiveness of the fusion module embedded in our network. Our code is available at: https://github.com/RCVLab-AiimLab/crowd_counting.

cs.CV↗

arXiv2021

The Hyperbolic Bloch Equations of General Relativity

New equations are derived which describe the evolution in curved spacetime of null geodesics with non-zero (complex) shear $σ$ and twist $ω$ rates resembling Grishchuk's squeezed states evolution equations from inflationary cosmology. A ``squeeze" angle $ϕ$ (obtained from the direction of the major axis of the elliptical cross section of the congruence and the direction of the shear rate), an ellipse axis ratio parameter $w$ and a rotation angle $v$ are the primary variables. Interpreting $ϕ$ as a polar angle and $w$ as a radial distance, we obtain a mapping to points on the upper sheet, $H_{2}^{+}\,,$ of a two-sheet hyperboloid, establishing the connection between gravitational optics and hyperbolic geometry. Points on $H_{2}^{+}$ trace out paths evolving according to hyperbolic Bloch equations, similar to the optical Bloch equations, which can also be represented as a Schrödinger-like equation with a non-Hermitian Hamiltonian. A single vector equation on $H_{2}^{+}$ describes the precession of hyperbolic Bloch vectors about a rotation or birefringence vector on $H_{2}^{+}\,,$ analogous to the precession of Bloch vectors on the Bloch sphere or Stokes vectors on the Poincaré sphere. Tidal gravitational effects and a non-zero twist $ω$ contribute to the precession of hyperbolic Bloch vectors.

gr-qc↗